Article, 2023
Artificial intelligence for predictive biomarker discovery in immuno-oncology: a systematic review
Annals of Oncology,
ISSN
1569-8041,
0923-7534,
Volume 35,
1,
Pages 29-65,
10.1016/j.annonc.2023.10.125
Contributors
Prelaj, Arsela
0000-0002-3863-088X
[1]
[2]
[3]
Miskovic, Vanja
0000-0003-1475-0243
[2]
[3]
Zanitti, Michele
[4]
Trovò, Francesco
[2]
Genova, Carlo
0000-0003-3690-8582
[5]
[6]
Viscardi, Giuseppe
0000-0003-4473-9387
[7]
Rebuzzi, Sara Elena
0000-0003-0546-6304
[6]
[8]
Mazzeo, Laura
0000-0001-9226-8861
[2]
[3]
Provenzano, Leonardo
[3]
Kosta, Sokol
0000-0002-9441-4508
[4]
Favali, M
[2]
Spagnoletti, Andrea
0000-0002-5293-1849
[3]
Castelo-Branco, Luis
[1]
[9]
Dolezal, J
[10]
Pearson, A T
[10]
Lo Russo, Giuseppe
0000-0003-3224-2728
[3]
Proto, Claudia
0000-0003-0287-9787
[3]
Ganzinelli, Monica
0000-0002-0526-0835
[3]
Giani, Claudia
[3]
Ambrosini, Emilia
0000-0002-6527-0779
[2]
Turajlic, Samra-
0000-0001-8846-136X
[11]
Au, Lewis
0000-0001-5877-8657
[12]
[13]
[14]
Koopman, M
[1]
[15]
Delaloge, S
[1]
[16]
Kather, Jakob Nikolas
0000-0002-3730-5348
[17]
de Braud, F
[3]
Garassino, M C
[10]
Pentheroudakis, George E
[1]
Spencer, C
[11]
Pedrocchi, Alessandra Laura Giulia
0000-0001-9957-2786
[2]
Affiliations
- [1]
European Society for Medical Oncology
[NORA names:
Switzerland; Europe, Non-EU; OECD];
- [2]
Politecnico di Milano
[NORA names:
Italy; Europe, EU; OECD];
- [3]
Fondazione IRCCS Istituto Nazionale dei Tumori
[NORA names:
Italy; Europe, EU; OECD];
- [4]
Aalborg University
[NORA names:
AAU Aalborg University; University; Denmark; Europe, EU; Nordic; OECD];
- [5]
Ospedale Policlinico San Martino
[NORA names:
Italy; Europe, EU; OECD];
(... more)
- [6]
University of Genoa
[NORA names:
Italy; Europe, EU; OECD];
- [7]
University of Campania "Luigi Vanvitelli"
[NORA names:
Italy; Europe, EU; OECD];
- [8]
Ospedale San Paolo
[NORA names:
Italy; Europe, EU; OECD];
- [9]
Universidade Nova de Lisboa
[NORA names:
Portugal; Europe, EU; OECD];
- [10]
University of Chicago
[NORA names:
United States; America, North; OECD];
- [11]
The Francis Crick Institute
[NORA names:
United Kingdom; Europe, Non-EU; OECD];
- [12]
Peter MacCallum Cancer Centre
[NORA names:
Australia; Oceania; OECD];
- [13]
Royal Marsden NHS Foundation Trust
[NORA names:
United Kingdom; Europe, Non-EU; OECD];
- [14]
University of Melbourne
[NORA names:
Australia; Oceania; OECD];
- [15]
Netherlands Comprehensive Cancer Organisation
[NORA names:
Netherlands; Europe, EU; OECD];
- [16]
Institut Gustave Roussy
[NORA names:
France; Europe, EU; OECD];
- [17]
TU Dresden
[NORA names:
Germany; Europe, EU; OECD]
(less)
Abstract
BACKGROUND: The widespread use of immune checkpoint inhibitors (ICIs) has revolutionised treatment of multiple cancer types. However, selecting patients who may benefit from ICI remains challenging. Artificial intelligence (AI) approaches allow exploitation of high-dimension oncological data in research and development of precision immuno-oncology.
MATERIALS AND METHODS: We conducted a systematic literature review of peer-reviewed original articles studying the ICI efficacy prediction in cancer patients across five data modalities: genomics (including genomics, transcriptomics, and epigenomics), radiomics, digital pathology (pathomics), and real-world and multimodality data.
RESULTS: A total of 90 studies were included in this systematic review, with 80% published in 2021-2022. Among them, 37 studies included genomic, 20 radiomic, 8 pathomic, 20 real-world, and 5 multimodal data. Standard machine learning (ML) methods were used in 72% of studies, deep learning (DL) methods in 22%, and both in 6%. The most frequently studied cancer type was non-small-cell lung cancer (36%), followed by melanoma (16%), while 25% included pan-cancer studies. No prospective study design incorporated AI-based methodologies from the outset; rather, all implemented AI as a post hoc analysis. Novel biomarkers for ICI in radiomics and pathomics were identified using AI approaches, and molecular biomarkers have expanded past genomics into transcriptomics and epigenomics. Finally, complex algorithms and new types of AI-based markers, such as meta-biomarkers, are emerging by integrating multimodal/multi-omics data.
CONCLUSION: AI-based methods have expanded the horizon for biomarker discovery, demonstrating the power of integrating multimodal data from existing datasets to discover new meta-biomarkers. While most of the included studies showed promise for AI-based prediction of benefit from immunotherapy, none provided high-level evidence for immediate practice change. A priori planned prospective trial designs are needed to cover all lifecycle steps of these software biomarkers, from development and validation to integration into clinical practice.
Keywords
AI approaches,
AI-based methodologies,
AI-based methods,
AI-based prediction,
Pathom,
algorithm,
analysis,
approach,
article,
artificial intelligence,
benefits,
biomarker discovery,
biomarkers,
cancer,
cancer patients,
cancer types,
changes,
checkpoint inhibitors,
clinical practice,
complex algorithms,
data,
data modalities,
dataset,
deep learning,
design,
development,
digital pathology,
discovery,
efficacy prediction,
epigenome,
evidence,
exploitation,
genome,
hoc analysis,
horizon,
immune checkpoint inhibitors,
immuno-oncology,
immunotherapy,
inhibitors,
integration,
intelligence,
learning,
lifecycle,
lifecycle steps,
lung cancer,
machine learning,
markers,
melanoma,
method,
methodology,
modalities,
molecular biomarkers,
multimodal data,
multimodality,
multiple cancer types,
non-small-cell lung cancer,
novel biomarkers,
oncological data,
original articles,
pan-cancer study,
pathology,
patients,
peer-reviewed original articles,
post,
post hoc analysis,
power,
practice,
practice change,
precision immuno-oncology,
prediction,
predictive biomarker discovery,
predictive of benefit,
prospective study design,
prospective trial design,
radiomics,
real world,
research,
review,
revolutionised treatment,
software,
standard machine learning,
steps,
study,
study design,
systematic literature review,
systematic review,
transcriptome,
treatment of multiple cancer types,
trial design,
type,
use,
validity,
widespread use
Funders
Data Provider: Digital Science